20080819 ARB Jump Start
Architecture Board
20080819 August out-of-cycle Day
Contents
- 1 Tuesday Morning 0800 - 1200
- 2 Tuesday 1300 - 1700
- 2.1 Summary from above (30 minutes)
- 2.2 ISCS (45 minutes)
- 2.3 OHT Modeling project for HL7 - Lessons Learned (Rogers) (60 minutes)
- 2.4 After break discussion of presentations
- 2.5 tbBAM - Structure of HL7 delivers support for customers to craft ISCS
- 2.6 Empower the feedback loop from implementation back to specification development.(30 Minutes )
Tuesday Morning 0800 - 1200
Attendance
Health Level Seven, Inc.| Name | Role | Affiliation | |
| Cecil, Lynch | ArB | ontoreason LLC | clynch@ontoreason.com |
| Charlie, Mead | ArB | Booz Allen Hamilton | charlie.mead@booz.com |
| Curry, Jane | ArB | Health Information Strategies | janecurry@healthinfostrategies.com |
| Julian, Tony | ArB | Mayo Clinic | ajulian@mayo.edu |
| Koisch, John | ArB | Boz | koisch_john@bah.com |
| Larsen, Ed | guest | HITSP | erlarsen@erlinc.com |
| Mulrooney, Galen ??? | guest | VHA | galen.mulrooney@med.va.gov |
| Orvis, Nancy | ArB | DOD | nancy.orvis@tma.osd.mil |
| Parker, Ron | ArB | CA Infoway | rparker@infoway-inforoute.ca |
| Quinn, John | ArB | Health Level Seven, Inc. | jquinn@HL7.org |
| Rogers, Rich | guest | IBM | rrogers@us.ibm.com |
| Shakir, Abdul-Malik | ArB | Shakir Consulting | ShakirConsulting@cs.com |
| Wrightson, Ann | guest | NHS | ann.wrightson@wales.nhs.uk |
| Yongjian, Bao | ArB | GE Healthcare | yongjian.bao@med.ge.com |
Meeting was called to order at 8:00am by John Koisch.
John went through the deck to date.
Review of Bernd Bobel's Document (30 minutes)
John presented Bernd's document that intersucts RUP, HL7, RM-ODP into a singe document. It is aligned with the DoD document at the Generic Componet Model. Agreement was reached to invite Bernd to the ArB Meeting in Vancouver.
John reviewed the 'bulls-eye' picture, and the intent of the ARB to help get down to the level needed. John reviewed the SAEAF and invited comments from the group.
Artifacts for SOA
- (I think this is a bit of a misnomer, esp in light of BB's work - this is the intersection of MDA, SOA, RM-ODP, a software engineering process, and HL7)
John presented the NCI work on Conceptual and Functional Apecifications.
- It was stated that if a team is using RUP, they have the basis of a SOA specification.
- In the "Information viewpoint", "the specification will support the static and underlying state models for the ____ Analysis Model."
- At the "Computational viewpoint", "the specification will support these groups of operations and their business context." This includes Rim specifications. "...Dependencies must be manifested in othere viewpoint's specifications, specially engineering. The conformance assertion is stated as a dependency on a function or functional profile."
Constraint model:
| RM-ODP Viewpoint | Constraint | Pattern | Mostly Information VP |
| +Enterprise +Information -Computation | Analysis Model | DAM | Blueprint |
| -Enterprise +Information +Computation -Engineering | Conceptual Design | DMIM, CIM, RMIM | logical platform independent |
| Information+Computation+Engineering xTechnology | Implementable Design (ITS, XSD, DDL) | LIM | Platform bound ( one man's platform is anothers logical model) |
| x Implementation | _ | Technology | _ |
Note: The interpretation of "platform" depends on your (non-rm-odp) viewpoint.
Specification Constraint Pattern
Each of the artifacts are expressed in a testable and verifiable way.
Transformation and Transition need to be understood - at both the implementation level as well as the conceptual design level, e.g. constraints on coded data-types.
- We need to decompose the business context - work product of blueprint.
15 Minutes Break @10:45
Review of FW Presentation(75 Minutes )
- How SOA artifacts are diffferent, how they map to the FW, how they map to the FW
- Traceability from Requirements to Design to Implementation (Traceability across artifacts)
- 15 Minutes - Break
Cross Functional Teams and SOA Artifacts (90 Minutes)
Real World Example: NCI COPPA Information Model work / Constraint Pattern (30 minutes)
John Koisch: John discussed the NCI COPPA analysis model - it was built for communication and education, not for development. Classes are flattened, relationships dis-ambiguated. COPPA - If you look at the HSSP spec Alan Honey et. al. wrote the EIS spec - which can be instantiated as a service, as well as used as a pattern. COPPA is the horribly misnamed. Cooreleation -covers the coorelations between the services. Organization Person Protocol Abstraction
We need to be able to support changing definitions over time - by providing provenance of the data.
Adjourned at 12:30 Tony 16:24, 19 August 2008 (UTC)
HDF is not very affected by services per se (60 minutes)
- need to show in S4 how unique SOA Artifacts support different aspects of the HDF, are created by different groups, etc.
Tuesday 1300 - 1700
Resumed Tony 17:46, 19 August 2008 (UTC)
Summary from above (30 minutes)
- Different groups need to expose XYZ artifacts to appropriately utilize/ support / give back to the HL7 Capability list. It is a QA issue when standards are in process.
In q1&2 we discussed the specification constraint patterns - everyone was comfortable with the results of the discussion.
ISCS (45 minutes)
John offered the NIH_LSACT_Specification_Methodology_Document as an example of the specification framework.
The box replaces all but the out two rings of the 'bulls-eye' diagram. The outer ring text is TSC, Org souce The next ring is ArB Developmet Governence Procedures The inner box id HL7 Soa Specs. The box contains the following text/Artifacts: Comformance and compliance RM-OPD UP Assumptions Assertions
Assumptions:
- RIM
- ADT
- CDA
- Messaging
- BVocab
- EHRS-FM
- COntent Models
- More - Payload message Contene(wrappers, etc).
New TOC - See SAEAF (From original agenda)
- Detailed Structure
- Relation to Conformance Statements
- Ties from Groups
OHT Modeling project for HL7 - Lessons Learned (Rogers) (60 minutes)
Rich Rogers presented the Open Health Tools(OHT) project File:Modeling-mdt20080820.zip.
Rich was given a list of Work Groups who might benefit from help.
ARB feels that this fits in "Next Steps."
Discussion was held to who the right group is for feedback. There were no conclusions. (14 people, 30 opinions :-})
Break! Tony 20:01, 19 August 2008 (UTC)
Break
After break discussion of presentations
Discussion was held updating the SAEAF
Discussion was held concerning who/what needs communication: Who: TSC, MnM, InM, SD Division, Vocab, EHR, SDO's, HSSP, Education, Publishing, Tooling.
ArB will meet (In vancouver) Sunday Q1, Q2 Tuesday Q4 Wednesday Q4 Thursday Q3, Q4
There should be communications
- before the WGM to
- SOA Work Group
- InM
- Technical Support Svcs SD(Education, Publishing, Tooling) (28th).
*During WGM
- TSC - Sat, Monday PM
- MnM - Th, Q5
- InM - Th, Q2
Invite Sdocs, Vocab, EHR Thursday Q3.
- After WGM
- SDO's - OMG, etc.
Tony 21:23, 19 August 2008 (UTC)
tbBAM - Structure of HL7 delivers support for customers to craft ISCS
- 30 Minutes - Tooling - It is not sufficient to simply rely on automated tooling (but we should def leveraged where we can).
- 30 Minutes - Capacity and Capability - We need to identify the "how to" as well as the phase transitions that are _easier_ and properly granulated.
15 Minutes Break
Empower the feedback loop from implementation back to specification development.(30 Minutes )
== Roles of HL7 Groups (60 Minutes) (including ArB) in
- Conformance and Compliance
- Process Mentoring
- QA Tony Julian
